OpenAI Releases Sora as Hollywood Struggles with AI March Into Industry

OpenAI has released Sora, its busy generative AI tool capable of creating hyper-realistic videos.

In a post issued on Monday OpenAI said the text-to-video generator is accessible for public use. An account must be created to use Sora, although sign-ups have been temporarily disabled due to heavy traffic.

The public launch of the tool comes as the entertainment industry grapples with implementing technology potentially capable of reducing production costs. Mainstream adoption in Hollywood has been slow but steady, with Lionsgate in September announcing a partnership with Runway in a deal that will see the New York-based AI startup train a new generative AI model on corporate content that will be used to help with the behind-the-scenes production processes. This was followed by James Cameron joining the board of Stability AI – the creator of Stable Diffusion, an image and video focused model that is among those being closely watched by many in Hollywood, especially in the visual effects industry – in a major coup for the company.

Videos up to 1080p resolution and 20 seconds long can be created in widescreen, portrait or square aspect ratio, according to the blog post. Users can upload their own assets to extend, remix and mix videos in addition to generating brand new content with a text prompt.

OpenAI said the tool will be included with existing ChatGPT Plus and Pro accounts at no extra cost. During the subscription, users can generate up to 50 videos in 480p resolution or fewer videos in a higher resolution.

Last month, OpenAI said ChatGPT now sees more than 200 million weekly active users. It plans to introduce ChatGPT Pro, a new $200 monthly subscription tier that includes unlimited access to OpenAI o1, GPT-4o and Advanced Voice mode. It will continue to offer a Plus tier for $20 per month, which includes early access to new features.

Sora was initially limited to a pool of security testers who reported vulnerabilities related to misinformation and bias, and visual artists, designers and filmmakers who provided feedback on improvements

A small group of artists who tested Sora in November leaked access to the tool in protest of the company’s treatment of them as “free bug testers” and “PR puppets.” In an open letter, they stated, “Hundreds of artists provide unpaid labor through bug testing, feedback and experimental work to the program of a company valued at $150 billion.”

While there are legal and labor protections, text-to-video tools are expected to have major applications in areas such as visual effects and animation. Some industry professionals have already adopted the technology in their workflows.

A study surveying 300 executives across Hollywood, released earlier this year, reported that three-quarters of respondents indicated that AI tools supported the elimination, reduction or consolidation of jobs in their companies. Over the next three years, it is estimated that nearly 204,000 positions will be adversely affected. Sound engineers, voice actors, concept artists and visual effects workers were cited as particularly vulnerable.

OpenAI no longer discloses the data sources used to train its systems. Artists, writers and publications have sued the Sam Altman-led company, alleging it illegally stole their work without consent and compensation. Courts ruling against fair use—a legal doctrine that allows the use of copyrighted works without a license—could have major implications for the AI ​​leader.